Hiltonol (poly-ICLC) is a stabilized dsRNA therapeutic viral mimic or 'danger signal' that activates multiple elements of innate and adaptive immunity. It is a standalone immunomodulator, but when properly combined with antigen it generates a comprehensive Th-1 weighted immune response best suited for antiviral and antitumor action. This project will combine Hiltonol with cancer patients' own tumor antigens in-situ to generate a comprehensive systemic immune response against the tumor and its metastases. Objectives: 1) To determine the safety of intratumoral (IT) Poly-ICLC. 2) To explore efficacy as manifested by irRC defined disease control and progression-free survival at 26 weeks. Secondary objectives are overall survival and immunologic response in blood and tumor biopsy, including immunoscore. Indications/Tumor groups: 1) Unresectable stage 3b, 3c or IV, M1a or M1b melanomas that are accessible to IT injections 2) Unresectable head and neck cancers that are accessible to IT injections 3) Sarcomas that are accessible to IT injections 4) Squamous cell carcinoma of the skin accessible to IT injections Study Design: A two-stage adaptive design Phase II clinical study with a pilot segment and an extension segment for the single tumor type showing the best clinical responses in the pilot phase. Pilot (SBIR phase I segment): Ten volunteer patients will be treated per group, for a total of 40 patients in the initil pilot phase of the study. Treatment consists of two cycles of intratumoral (IT) Hiltonol priming plus intramuscular (IM) Hiltonol(R) maintenance. One accessible tumor site will be targeted for initial IT injections of poly-ICLC thrice weekly for 2 weeks, followed by IM maintenance twice weekly for six weeks in each cycle. A second cycle will be followed by a 6-week no-treatment rest period to allow for evaluation of tumor response at week 26 in the absence of inflammation. Extension (SBIR phase II segment): Based on the safety and clinical response for patients in each tumor group during the pilot phase of the study, one of the 4 groups showing IRRC-defined stabilization or regression of disease at 26 weeks in at least two of the initial ten patiets will be selected for further study under this proposal. Appropriate, indication-specific statistics will be applied and an estimated 60 additional patients will be accrued to that group, for an estimated total of 70 patients in that group. Accrual targets may be further modified depending on the extent of response, other data collected, and results of a planned industry meeting with FDA. Survival and PFS will be estimated using Kaplan-Meir curves.